Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Havacılık ve Uzay Mühendisliği Bölümü, Türkiye
Tezin Onay Tarihi: 2016
Tezin Dili: İngilizce
Öğrenci: Sema Karahan
Danışman: ALİ TÜRKER KUTAY
Özet:The aim of this thesis is to design a fault detection and accommodation (FDA) algorithm for air data systems, and offer a new calibration algorithm for five-hole probes (5hp) that can tolerate one port blockage and continues to give correct data. Air data systems use 5hp to measure the magnitude and direction of airspeed. However, five-hole probes are very vulnerable to obstruction. Numerous fatal accidents happened because of probe blockages. Hardware redundancy is applied as a precaution for air data system failures, but this does not provide a solution against common-mode failures. Analytical redundancy arises as a solution, and different methods have been utilized for FDA. In these methods, wind data estimation is critical for the detection performance. Wind data are either taken from another source or estimated. In the scope of this thesis, a signal based fault detection system that does not require any wind data input is developed. Then, a model based fault accommodation algorithm is implemented. The algorithm is tested under various wind and turbulence conditions. At each trial fault is detected within 0.5 seconds, and accommodated. Current calibration algorithms for 5hp do not allow the isolation of a faulty port reading, and even if one port is blocked, they fail to give correct outputs. Contrarily, the new calibration algorithm eliminates the erroneous data from the algorithm, and continues with healthy data. This approach enhances the operating conditions of 5hp.